Metraipolis › archive › dc_bac044f127f7

Diagnosing the Fact-Grounding Gap in Multi-Hop Question Answering — arXiv:2609.17043

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.17043
License (as archived)
CC0-1.0
Archived
2026-10-15T18:38:47.230Z
SHA-256 of the archived text
c7f01b4ca32cd59dc9c9c679fa0afdc7db36b4b26e991fdd83dc54303674b13b
Stable URL of this page
https://metraipolis.com/documento/dc_bac044f127f7

How to cite

Diagnosing the Fact-Grounding Gap in Multi-Hop Question Answering — arXiv:2609.17043 — arXiv (en), https://arxiv.org/abs/2609.17043 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bac044f127f7 (Metraipolis archive dc_bac044f127f7, sha256 c7f01b4ca32cd59dc9c9c679fa0afdc7db36b4b26e991fdd83dc54303674b13b)
BibTeX
@misc{dcbac044f127f7,
  title = {Diagnosing the Fact-Grounding Gap in Multi-Hop Question Answering — arXiv:2609.17043},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.17043},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bac044f127f7; sha256 c7f01b4ca32cd59dc9c9c679fa0afdc7db36b4b26e991fdd83dc54303674b13b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.17043. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multi-hop question answering requires combining information from multiple documents to answer complex questions. These systems have grown increasingly capable, yet when they fail, the error is typically attributed to not finding the right documents. Whether this holds at the level of individual reasoning steps remains largely unexamined. We investigate this across three standard multi-hop QA benchmarks and find that failures decompose into two distinct modes: retrieval failures, where the needed passage was not retrieved, and extraction failures, where the passage was retrieved but the needed fact could not be extracted - a phenomenon we term the fact-grounding gap. Extraction failures account for nearly half of all per-hop deficiencies and are invisible to standard retrieval metrics. They remain unresolved by every retrieval intervention we test, establishing a ceiling for retrieval-only improvements. The gap's severity varies across benchmarks and question types, but extraction failures appear on every dataset we measure. Our findings reveal that retrieval failures and extraction failures are fundamentally different bottlenecks requiring different solutions - a distinction absent from current evaluation practice.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-16 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2609.17043 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T18:03:09.718Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).